A Supervised Ensemble Learning-Based Approach to Mitigate SQL Injection Attack on Smart City Data
AsifIqbal Sirmulla, M Prabhakar · International Journal of Engineering Trends and Technology · 2023
Currently, the use of internet-based applications and technologies has grown drastically. Due to this growth, a massive amount of data is generated and stored on webservers. This increased use of these technologies faces several challenges, such as data vulnerability, security threats and data privacy. SQL-based programming language is widely adopted to access these data due to its simple and efficient use. However, the attacks can use the SQL-based query injection method, where they can insert malicious queries, posing serious threats to the server. Several techniques have been presented in the past, such as blacklisting and rule-based detection methods, but these methods fail to detect SQL injection attacks due to their diversity in input queries. Thus, currently, machine learning-based schemes have gained massive attention in this field, where supervised and unsupervised methods are widely employed. However, the varying nature of SQL queries demands a stable architecture. This work presents a machine learning-based approach for SQL injection attack detection by introducing an ensemble machine learning approach where SVM, NN, discriminant and random forest classifiers are employed. The experimental analysis shows that the average accuracy performance is achieved as 95.88%, 96.23%, 96.61%, 95.11%, and 99.30% using KNN, Discriminant classifier, Naïve Bayes, SVM, and proposed ensemble classification.